Jan-Hendrik Menke

login after locking screen results in empty workspace

3

after having locked the screen, sometimes (probably in 20 % of cases), i have to login twice, my first login is ignored, and then the login screen is shown again....

automatic detection of multioutput datasets

10

## What does this PR do? Slight variation of #903 depending on the shape of the target array. ## What are the relevant issues? #971

Recently, Optuna has emerged as a great auto ml tool. Could you add it to your benchmark?

Add save/load functionality and fix issue with sorted

5

- Implemented simple save/load of a fitted encoder by using pickle - sorted was removed in a commit, which let the tests fail due to random orders of the column_mapper...

DuckDB file keeps growing on data overwrite

13

### What happens? I am connecting to DuckDB from Julia using DuckDB.jl In an initial run, a lot of data is crawled and stored in different tables. On subsequent runs,...

from_json / to_json does not throw exceptions

9

Suppose i save my file {"x":3} as "net.json". Now i can use pp.from_json("net.json") without an error despite it not being a pandapower network. It goes through the decoder without errors....

Callback change from 1.10.0 to 1.11.0 has too much overhead

9

I am solving an optimization problem in Julia using JuMP.jl and HiGHS.jl After updating from HiGHS.jl 1.17.0 to 1.18.1 (which internally uses HiGHS 1.11.0 instead of 1.10.0), the solver needs...

Updating to 1.18.0 (or 1.18.1) slows down optimization considerably

19

Continuing here from the [original issue in the HiGHS github repo](https://github.com/ERGO-Code/HiGHS/issues/2466). Basically since updating HiGHS.jl from 1.17.0 the solve time is much slower across multiple machines. Cross testing with the...